Satisfaction, research productivity, and socialization in doctoral students: Do teaching assistantship, research assistantship and the advisory relationship play a role?

Abstract

Funding to doctoral students in the form of research and teaching assistantships help students become independent scholars and complete their programs. Insufficient funding, unforeseen financial obligations, and debt can discourage students from completing their programs in a timely fashion. However, supervisors may play an influential role in supporting doctoral student socialization and growth towards research autonomy. Girves and Wemmerus's (1988) doctoral student degree progress theoretical model claimed that financial assistance and students' perceptions of faculty support are key predictors of doctoral students' progress. They also proposed that students' satisfaction with their department, sense of isolation, and engagement in their programs may explain (or mediate) these associations. The aim of this study was to investigate how supervisor support and financial assistance influence doctoral students' satisfaction with their programs as well as their academic and social engagement. Using data from the Canadian Graduate and Professional Student Survey (n = 18,822 doctoral students), we evaluated a model of doctoral student productivity (i.e., publications and conference presentations) with teaching and research assistantships as key predictors and student satisfaction as mediators. We also examined how supervisor support and financial assistance relate to students' academic and social life. The results showed that teaching assistantships are negatively associated with general satisfaction and program satisfaction but positively associated with social involvement. Moreover, research assistantships were significantly and positively associated with social involvement and progress. Supervisor support was positively associated with doctoral students' general satisfaction, program satisfaction, and social involvement. The findings support Girves and Wemmerus's theory and highlight the importance of faculty support for doctoral student success. (258/300).

Highlights

• •Doctoral students with more financial aid graduate and conduct more research.
• •Research assistantships positively predict socialization, presentation, and publication.
• •Teaching assistantships negatively predict satisfaction in doctoral students.
• •The supervisory relationship influences doctoral students' mental health.
• •Doctoral supervisors positively influence student satisfaction and socialization.

1. The impact of funding on doctoral student success

Inadequate financial support is a major contributor to student dropout from doctoral programs. Student funding comes in different forms, including internal awards from departments in the form of research and teaching assistantship positions and external awards from provincial, federal, and international councils. While doctoral students with better financial support are more likely to graduate [[1], [2], [3]], and conduct more research [4], many federal and provincial grants and loans are unavailable to graduate students. Changes to funding policies, rising tuition fees, and availability of bank loans all contribute to doctoral student attrition because they have made doctoral studies more accessible and more expensive at the same time [5]. Tuition and fees are major expenses for many doctoral students [6]. The rising cost of graduate education causes financial stress and negatively affects doctoral students’ focus and efficiency in completing their program [2,[7], [8]].

Most doctoral programs financially support doctoral students through part-time teaching, and research assistantship positions, but there are considerable differences between disciplines in the generosity of these awards [9]. These offerings are sometimes mandatory and teach students how to construct a course and deliver a lecture [10]. However, teaching and research assistantships are mainly designed to fulfill departmental needs instead of enhancing doctoral training [11]. Similar to the doctoral student degree progress theoretical model [12], studies found that research and teaching assistantship positions facilitate social connections among doctoral students [13,14] and with the faculty and staff in their departments [15,16] and reduce the need to seek outside employment.

Research has also found that types of funding affect doctoral students differently. Zhou and Okahana [17] found that research assistantships predicted shorter times-to-degree, while teaching assistantships negatively predicted time-to-degree. However, the effects of funding sources could also differ between disciplines and programs. For example, STEM doctoral programs with more faculty members with grants and awards were more likely to have higher completion rates than non-STEM programs [17]. Furthermore, students in social sciences, education, and humanities who had larger graduate loans have a shorter time-to-degree than non-borrowers [18]. The student's stage of doctoral studies is another factor that should be considered because the availability of funding support changes over the course of the program. A study by Sowell et al. [19] showed that doctoral students experienced more financial problems when they entered the candidacy stage because most fellowships are restricted to students in the first three years of the program.

Financial strain is a significant barrier to doctoral student completion [3,8,20] and predicts dropout [21,14,22]. Financial support, on the other hand, has positive effects on doctoral students' retention, improves the likelihood of degree completion [13,14,19], and reduces time to completion [17,18]. The benefits of funding support are not limited to doctoral retention and shorter times to completion. Research funding is also associated with students’ productivity and scientific impact. A study by Larivière [3] revealed that funded students had approximately 25% more publications than unfunded doctoral students. Moreover, the type of assistantship affects how students integrate with their faculties [23]. Doctoral students who receive teaching and research assistantships spend more on campus and interact more with faculty members and peers compared to students who do not receive this support [23]. In other words, financial support to doctoral students facilitates opportunities to socialize in their departments [15,24].

Although published studies to date have focused on graduate students' retention, there is a lack of research on the relative importance of different sources of financial support on graduate students’ persistence in their programs.

2. The impact of supervisor on doctoral student

Supervisors have a powerful influence on doctoral student socialization [23]. They shape and structure the learning experience for students [23,25] and directly and indirectly foster a culture for doctoral students to transition to research autonomy [25]. They are the principle factor that determines student collaboration and involvement in different research opportunities [24,26]. Since students experience various psychological stressors and mental health challenges at each stage of a doctoral program, the student-supervisor relationship can be critical throughout these stages [7,23]. Supervisors can buffer the stress of graduate school and facilitate the socialization process, especially when students lack interaction with other faculty members and peers [13,27]. Therefore, having a positive relationship with supervisors carries great importance for students because it affects various opportunities to integrate academically [24,28].

Supervisors also influence students' satisfaction with their program [7], writing efficacy [29], mental health [30], well-being [31], research self-efficacy [32], and persistence [33,34]. Supervisors influence students’ understanding of graduate school and its requirements, selection of dissertation topics, and future career opportunities [23]. According to McAlpine and McKinnon [31], more than 50% of students rely on their supervisors as primary resources on all academic matters—not only research projects. Therefore, finding a supportive supervisor is crucial for new doctoral students.

Good supervisors support students in meeting degree requirements and departmental deadlines and preparing for qualifying exams [33]. Qualitative studies show that the supervisory relationship strongly influences doctoral students' emotions [31]. An unsatisfactory relationship between supervisor and advisee contributes to student dropout [23,31]. Indeed, students’ perceptions of their interactions with faculty supervisors affect their feelings of connectedness and belongingness [35] and help them perform academically better than students who lack these feelings of connectedness and belongingness [36].

Students value the quality of support, feedback, level of trust, and communication in their relationship with their supervisors [35]. This support can be emotional, informational, or practical and can be assessed with respect to actual support received or perceived support one believes to be available. All these types of support promote student well-being [37], and contribute to a better supervisory relationship and improved educational experiences [38]. Most doctoral students expect their supervisors to be open with them about disciplinary norms, value the pursuit of knowledge, and define what it means to be successful [24].

Numerous studies have examined the characteristics of doctoral supervisors and described important characteristics to look for when selecting a supervisor [[39], [40], [41]]. These qualities include reputation in research, the convergence of interest, and being time-conscious [39,41], a good communicator [35,39,42], and supportive [38]. Ray and Marakas [39] ranked ten criteria for supervisor selection: a) committed and involved in graduate education, b) able to defend and support the student in contentious situations, c) has a strong scholarly reputation, d) is time conscious, e) shares research interests with the student, f) willing to support the student's career development, g) flexible about using alternative approaches in research, h) warm interpersonal style, i) has a large network of collaborators within and outside the program, and j) has a record of successfully supervising doctoral students. Similarly, Rose [40] described communication skills (“My ideal mentor would communicate openly, clearly, and effectively”) and the provision of feedback (“My ideal mentor would provide honest feedback; both good and bad, about my work”) as the two main characteristics of an “ideal mentor.” Moreover, in a second study by Rose [40], integrity, guidance, and relationship were defined as three dimensions of an ideal mentor.

Doctoral students' mental health is substantially impacted by their supervisory relationship [43]. Given the central role of the supervisor in doctoral students' academic life, it is not surprising that perceptions of an unsatisfactory student-supervisor relationship contribute to doctoral student depression [43]. Several studies cited the supervisor as the primary determinant of students' academic progress. However, effective doctoral supervision consists of academic and personal support and can result in less anxiety and depression for students [30]; there exists a notable gap in research on the impact of supervisor support on the psychological well-being and satisfaction of doctoral students. Therefore, more research is needed to give insight into the impacts of both financial support and the supervisory relationship on doctoral students’ well-being.

To summarize, considering the high dropout rate of doctoral students and the abovementioned research on the determinants of student attrition, knowledge about the importance of different types of financial support in doctoral students' social life and their effects on the attrition of doctoral students is needed. Previous research also suggests that investments in financial aid should keep pace with increasing educational expenses to minimize financial pressure and remove financial barriers to program completion. Using data on a large sample of graduate students in Canada, this study tested three hypotheses based on Girves and Wemmerus's [12] doctoral student degree progress theoretical model.

Hypothesis 1

Financial support (i.e., teaching assistantship and research assistantship) is positively related to program satisfaction, general satisfaction, and social involvement.

Hypothesis 2

Supervisor support relates to higher program satisfaction, general satisfaction, and social involvement.

Hypothesis 3

Program satisfaction, general satisfaction, and social involvement mediate the associations of supervisor support and financial support with research productivity.

3. Methods

3.1. Present study

Girves and Wemmerus [12] doctoral student degree progress theoretical model

Identifies financial support and students' perceptions of faculty support as key predictors of doctoral students’ progress. They also suggest that these links are explained (or mediated) by student satisfaction with the department, student isolation, and student involvement in the program. We examined doctoral student progress (measured in research publications and presentations) using data from the Canadian Graduate and Professional Student Survey (CGPSS). Specifically, we tested a model that identifies teaching and research assistantships as financial supports, recognizes general satisfaction, program satisfaction, and social involvement as mediators, and the number of publications and presentations as progress (Fig. 1). The aim of the study was to examine how the academic and social lives of doctoral students in Canada are related to supervisor support and financial support.

Fig. 1.

Hypotheses model of the Study

Note. Hypotheses of the study based on the doctoral student degree progress theoretical model by girves, j. E., & wemmerus, V. (1988).

3.2. Participants

The 2016 survey cycle of the CGPSS provided the data for this study. The CGPSS is an ongoing cross-sectional survey of graduate students administered by the Canadian Association of Graduate Studies (https://cags.ca/cgpss). While the initial sample of the CGPSS data included doctoral students (34.5%), master's research students (MA Research; 32.4%), and master's professional students (MA Professional; 33.1%), only doctoral students were included in the current study. The sample included 18,822 doctoral students from 53 universities across Canada surveyed in 2016.

Doctoral students were asked to specify the current stage of their program, and 27.1% reported being in the coursework stage, 18.2% completed their coursework, 25.8% were in the comprehensive examination stage, 24.9% were in the dissertation stage, and 3.9% had defended their dissertation. The sample was composed of 53.1% of female students and 46.9% male students. The majority (78.2%) chose English as their first language, and the rest (21.8%) stated that French was their primary language. Relationship status was 32.6% not married, 35.8% married, and 14.8% living with partners. Whereas 46.8% mentioned that their primary reason for starting the doctoral program is a career in academia, 20% were not interested in working in academia, and 28% stated that they chose a doctoral program to satisfy their needs in their field. In terms of the year of study, 18% of participants were in the first year of the program, 20.1% in the second year, 20.2% in the third year, 17.3% in the fourth year, 13.4% in the fifth year, and 10.9% in the sixth year or above.

3.3. Independent variables

3.3.1. Financial support

Students were asked to specify if they had received any graduate research assistantship or graduate teaching assistantship during their program. Nearly half of the students reported they had not received a university fellowship (57.4%). A minority indicated that they received financial support through teaching assistantships (38.7%) and research assistantships (49.6%).

3.3.2. Supervisor support

The survey used a 14-item scale that measured supervisor support in a variety of mentoring activities. Response options ranged from 1 = strongly disagree to 4 = strongly agree (α = 0.96; M = 46.67, SD = 8.96). Scale items assessed different aspects like writing support (e.g., “My advisor was very helpful to me in writing a dissertation prospectus or proposal”), as well as availability (e.g., “My advisor was available for regular meeting”, “My advisor returned my work promptly”).

3.3.3. Dependent variables

3.3.3.1. Presentation and publication progress

We assessed doctoral students’ productivity and progress in doing research using three items that measured the frequency of presentations and publications (e.g., the number of occurrences of “Deliver any papers or present a poster at national scholarly meetings” and “Published as sole or first author in a refereed journal”). Responses ranged from 0 = zero time to 4 = more than four times was used to measure; α = 0.71 M = 3.14, SD = 3.14).

4. Mediators factors

4.1. Social involvement

The frequency of attending social activities on campus was used to measure doctoral students’ social involvement (i.e., social life). A 4-item scale ranging from 1 = never to 3 = frequently measured the number of social activities attendances (e.g., “Organized social activities within your advisor/research group”; α = 0.75; M = 7.70, SD = 1.98).

4.2. General satisfaction

To evaluate general satisfaction, a 5-item scale ranging from 1 = definitely not to 5 = definitely, was used to measure whether or not a doctoral student would choose to start over the program (e.g., “If you were to start your graduate/professional career again, would you select the same field of study”; α = 0.75; M = 19.70, SD = 3.83).

4.3. Satisfaction with the program

The quality of the coursework, interaction, and satisfaction with the program was assessed by a 13-item scale ranging from 1 = poor to 5 = excellent (e.g., “Overall quality of graduate level teaching by faculty”, “Quality of advising and guidance”; α = 0.90, M = 44.73, SD = 9.76).

4.4. Data analysis

RStudio version 4.1.2 was used for data management and analysis. Correlational analyses were conducted on all continuous variables, including supervisor support, general satisfaction, program satisfaction, social involvement, and progress. One-way analyses of variances (ANOVAs) were utilized to explore potential gender differences in the measured variables. Structural equation modelling (SEM) was used to examine whether general satisfaction, program satisfaction, and social involvement mediated the relationship between teaching and research assistantship and progress (i.e., publications and presentations) as well as the relationship between supervisor support and progress.

5. Results

5.1. Preliminary analyses

For descriptive purposes, we used one-way ANOVAs to identify gender differences in the measured variables. There were significant gender differences in program satisfaction F (1, 16,230) = 29.43, p < 0.001, supervisor support F (1, 15,390) = 59.05, p < 0.001, social involvement F (1, 14,629) = 79.16, p < 0.001 with males reporting more program satisfaction (M = 44.99, SD = 9.94), better supervisor support (M = 3.37, SD = 0.61), and better social involvement (M = 1.99, SD = 0.48), than female students (M = 44.15, SD = 9.71; M = 3.29, SD = 0.65; M = 1.92, SD = 0.48 respectively).

Correlational analyses were conducted of continuous variables (Table 1). Anticipated supervisor support was strongly and positively correlated with general satisfaction r (16,223) = 0.49, p < 0.001, satisfaction with program r (16,219) = 0.47, p < 0.001, and social involvement r (14,093) = 0.18, p < 0.001. However, supervisor support was only weakly correlated with progress r (16,123) = 0.10, p < 0.001.

Table 1.

Correlations between variables (n = 18,822).

	1	2	3	4	5
Supervisor support	_
General Satisfaction	0.49**	_
Program satisfaction	0.47**	0.64**	_
Social involvement	0.18**	0.17**	0.19**	_
Progress	0.10*	−0.02**	0.001	0.12**	_

Note. *p < 0.05. **p < 0.001.

5.2. Factor analysis and reliability

Following the preliminary analysis, an initial principal component analysis with Oblimin rotation identified a 6-factor solution based on the eigenvalue greater than 1 criterion [44]. In order for factors to be considered important, one factor was removed that had items less than |0.4| (meaning ≥ +0.4 or ≤ −0.4) [45]. Therefore, factor analysis was conducted again with the extraction method changed to fixed five factors. The Kaiser–Meyer–Olkin MSA was 0.959. The five dimensions explained a total of 59.27% of the variance among the items in the study. Bartlett's Test of sphericity proved to be significant (p < 0.001).

Five constructs (i.e., advisor support, general satisfaction, program satisfaction, social involvement, and progress), and two single items measuring teaching and research assistantship were included in the structural equation modelling. The model fit indices were included the comparative fit index (CFI), Tucker–Lewis Index (TLI), and root mean square error of approximation (RMSEA). A comprehensive CFA analysis including all study measures showed an acceptable fit of the data, χ² = 7261.65, df = 741, p < 0.001, CFI was 0.95, TLI was 0.95, SRMR was 0.04 and RMSEA was 0.04 (See Appendix A).

5.3. Structural equation modelling

Our mediational analysis used SEM to test the fit of the model shown in Fig. 2. The model shows general satisfaction, satisfaction with the program, and social involvement as mediators between supervisor support, research assistantship, teaching assistantship, and publications and presentations. Indices of goodness-of-fit to the data were acceptable, χ² = 4228.895, df = 748, p < 0.001, CFI was 0.95, TLI was 0.94, RMSEA was 0.04, and SRMR was 0.04. Teaching assistantships negatively associated with satisfaction with the program (β = −0.11, SE = 0.04, p = 0.001), but positively predicted social involvement (β = 0.29, SE = 0.05, p < 0.001). Supervisor support was the strongest direct predictor of general satisfaction and satisfaction with the program and was positively associated with general satisfaction (β = 0.78, SE = 0.07, p < 0.001), program satisfaction (β = 0.63, SE = 0.03, p < 0.001), and social involvement (β = 0.16, SE = 0.02, p < 0.001). Moreover, supervisor support indirectly predicted presentations and publications (progress) through social involvement, β = 0.02, SE = 0.00, p < 0.001. The results of indirect analyses suggest a positive association between teaching assistantships and progress mediated by social involvement, β = 0.03, SE = 0.01, p = 0.001. Research assistantships positively predicted social involvement (β = 0.14, SE = 0.04, p = 0.002). Also, holding research assistantships was indirectly associated with presentations and publications (progress), mediated by social involvement, β = 0.01, SE = 0.01, p = 0.013.

Fig. 2.

Structural Equation Model

Note. Structural equation model showing general satisfaction, satisfaction with the program, and social involvement as the mediators between supervisor support, research assistantship, teaching assistantship, and presentations and publications (progress). Standardized coefficients are shown on each path.

*p < 0.05, **p < 0.001.

6. Discussion

Drawing upon the existing literature, the present study examines the relationship between different types of support (i.e., teaching and research assistantship, and supervisor support) and doctoral students' satisfaction, social involvement, and academic progress. Previous research underlines the crucial role of financial support in doctoral student retention, graduation rates, and research productivity. Various forms of assistantships, encompassing both teaching and research roles, have been identified as substantial financial supports, concurrently fostering beneficial social connections among doctoral students. Although the effects of different types of financial support may vary, a consensus emerges in the literature emphasizing the detrimental impact of financial strain and the positive outcomes correlated with robust financial support. Moreover, the influence of supervisor support on doctoral students' academic lives has been explained in several studies. The quality of support, level of trust, and quality of communication within the student-supervisor relationship are frequently identified as critical elements. However, there is a noticeable lack of research studying how a supervisor's support and financial support influence doctoral students' social life and satisfaction. Therefore, our study aimed to test three hypotheses based on Girves and Wemmerus's [12] doctoral student degree progress theoretical model and that explore the relationship between financial support (i.e., research and teaching assistantship), supervisor support, and research productivity mediated by satisfaction (i.e., general and program satisfaction) and social involvement.

Our results are consistent with previous studies which show that socialization in doctoral students is positively influenced by research and teaching assistantship positions [13,14]. Financial support provides students with the means to attend departmental events, conferences, and workshops, which can create opportunities for networking and socializing with other students and faculty members. These activities help students feel more connected to their academic community and reduce feelings of isolation [46]. Financial support for doctoral students facilitates opportunities for socialization within their departments [15,24]. The study results show that financial support for doctoral students facilitates opportunities for social involvement. However, teaching and research assistantships differ in their links to program satisfaction and general satisfaction, probably due to the fact that teaching assistantships divert time and other resources away from students’ research and degree completion, especially during the dissertation stage [47,13]. As a result, this may explain why teaching assistantships lead to decreased general and program satisfaction among doctoral students who receive them.

The results also show that supervisors have significant influence on their doctoral students' general satisfaction, program satisfaction, and social involvement. Supervisors affect their students' academic progress both directly and indirectly by improving students’ well-being. These results also highlight the importance of fostering effective and collaborative relationships between supervisors and students. Supervisors foster a culture beneficial to doctoral students to help them achieve research autonomy [25]. They are the primary determinant of student collaboration and participation in various types of research [24]. Without proper communication, mutual goals, and academic and personal support, students are more prone to mental health issues that could result in performance declines and program withdrawal. Thus, maintaining a strong relationship with supervisors is critical for students since it influences several opportunities for academic integration [24]. These findings encourage students, supervisors, and faculty to initiate better relationships to help doctoral students manage the intense workload and eliminate the lack of socialization experienced during their doctoral training to more effectively dedicate their time to research and persist confidently.

The most significant finding from this study is that supervisor support has a greater association with mediators such as general satisfaction and program satisfaction, compared to the other two predictors, teaching and research assistantship. Supervisors have an impact on students’ perceptions of graduate school and its requirements, dissertation subject selection and quality, and future career opportunities [23]. Supervisors also have an effect on how satisfied students are with their program [41]. Consistent with previous research, we found support for hypothesis two— that a greater degree of support from supervisors may significantly boost doctoral student program satisfaction, general satisfaction with life, and social involvement.

Finally, hypothesis three stated that program satisfaction, general satisfaction, and social involvement mediate the associations of supervisor support and financial support with research productivity. The third hypothesis was not entirely supported. General and program satisfaction were not identified as significant mediators of the relationship between supervisor and financial support and research productivity. However, we did find that social involvement mediated the relationship between research assistantship and academic achievement among doctoral students. Increased financial assistance for students in the form of research assistantships resulted in increased social involvement and, consequently, more productivity.

6.1. Strengths and limitations

The strengths of the study include a large sample of 18,822 doctoral students, which provided a comprehensive understanding of the factors that influence doctoral student achievement. Additionally, this study evaluated a model of student productivity that included both social and structural factors. This model provides a comprehensive understanding of the factors that contribute to doctoral student success. The findings have practical implications for graduate education by providing insights into the determinants of doctoral student success and thereby informing efforts to support graduate studies.

Limitations of the study should also be taken into consideration when interpreting these results. The response rate to the CGPSS survey was also relatively low (32.3%), and the study relied on self-report questionnaires to assess supervisor support, social involvement, program satisfaction, and general satisfaction. The survey questionnaire omitted mental health assessments; consequently, this study did not include issues such as depression, anxiety, and stress. It should be noted that even though the organizational context of departmental and supervisor supports influence the process of socialization and psychological well-being, the institution-level characteristics were not measured in the study survey. In addition, well-being greatly influences research productivity and teaching and ultimately the quality of higher education [48].

Further research is needed to further investigate how supervisor and financial support influence doctoral students’ psychological well-being. More evidence on the consequences of other forms of financial support, such as scholarships and loans, is also needed. In addition to supervisor support and financial support, it would be interesting to study the function of social support networks, such as peer support groups and counseling services, in promoting doctoral students' well-being and academic achievement. In addition, future research might study the influence of these variables on the career outcomes of doctoral students, such as employment opportunities and job satisfaction. Overall, this is an important area of research that supports efforts to optimize performance and well-being of doctoral students.

7. Conclusion

The pathway to receive scholarships, like becoming a doctoral student, is competitive and based on academic merit, a high-quality research proposal, and leadership experience [49]. Doctoral students surveyed in the study by Nettles and Millett [50], reported that 44% were offered research assistantships, 60% were offered teaching assistantships, and 48% were offered fellowships. Fellowships are counted as the “top prize because they often cover all student expenses and ordinarily come with no work requirements. Research and teaching assistantships, however, which often require students to work with faculty on research projects or instructional activities, can be most valuable for their associations and the apprenticeships they provide to students in preparation for professional careers” (48, p. 74). We found that doctoral students with teaching and research assistantship support will experience less satisfaction with their programs and less desire to start over the same program. According to Lovitts [23], the attrition rates of students who held teaching assistantships and research assistantships were 24 and 17%, respectively. This might explain that lower satisfaction leads students to leave the doctoral program.

Our findings suggest that financial support could improve doctoral students' social and academic lives. Financial aid in the form of assistantships allows doctoral students to be more involved in academic tasks with their peers and faculty members, which is critical in doctoral student retention [18,51]. One of the responsibilities of graduate deans is to allocate financial resources to their students [23]. The type and amount of funding vary between disciplines [3] and, as shown in this study, have different effects on student outcomes. Therefore, financial aid and other departmental and institutional policies and practices should be carefully evaluated in their efficacy in improving doctoral students’ lives.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contribution statement

Samira Feizi: Conceived and designed the experiments; Performed the experiments; Analyzed and interpreted the data; Contributed reagents, materials, analysis tools or data; Wrote the paper.

Frank Elgar: Conceived and designed the experiments; Wrote the paper.

Data availability statement

The data that has been used is confidential.

Declaration of interest's statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article.